#stat432


Exam II, COVID-19

Date and Times

  • Start: Thursday, April 16, 12:01 AM (Champaign Time)
  • End: Thursday, April 16, 11:59 PM (Champaign Time)

Format

Exam II will not use the CBTF. Exam II will be a quiz on PrairieLearn.

You will be given two hours to complete Exam II within the 24 hour window defined above. However, you must start two hours before the end time of 11:59 PM, otherwise, you will only be able to work until 11:59 PM.

Like Exam I, Exam II will have multiple, but not unlimited attempts.

  • You will be given three attempts for checkbox questions. The first attempt will allow full credit. The next two attempts will allow half credit.
  • You will be given six attempts for numeric questions. The first three attempts allow full credit. The next three attempts allow half credit.

The question format will mirror that seen in the quizzes. Some quiz questions will reappear during the exams.

If you get a question incorrect, a new version will not be generated, however you will not see the submission hints. Follow-up attempts will have the same version of the problem. (This is different than the quizzes where you always get a new version. If a quiz question is reused for the exam, it may indicate that new variants will use a different seed, but no new variants will be generated on the exam.)

The exam is open books, notes, and internet.

Content

Exam 02 will cover the material from Quiz 06 to Quiz 10. Because of delays in the release of Quiz 09 and Quiz 10, the exam will focus on Quiz 06, Quiz 07, and Quiz 08. There will be at most two or three questions related to Quiz 09 and Quiz 10.

In addition to multiple attempts, there will be more questions available than are needed to obtain 100% on Exam 02. Point cutoffs to obtain full marks will be indicated in the directions of the exam, which you should read as soon as you begin the exam.

In particular, you should be able to:

  • Fit and validate classification models.
  • Cross-validate any metric for any model.
  • Calculate any of the binary classification metrics here: https://en.wikipedia.org/wiki/Confusion_matrix
    • You can use this page during the exam!
    • You should know the definitions of TP, FP, TN, FN

Academic Integrity

Given the situation, detecting cheating will be difficult. If we do detect cheating, that means it will have been an extremely flagrant violation and we will seek the maximum possible punishment which at minimum would be a failing grade for the course. PrairieLearn will ask you to accept an “honor code” at the start of the exam.

Technology

Because you are using your own computer for this exam, you should be sure that your R and RStudio are setup and ready to go.

R Packages

You should be sure to have the following packages installed:

  • tidyverse
  • caret
  • rpart
  • mlbench
  • glmnet
  • randomforest

Issues

Many technology issues can be solved by simply restarting RStudio.

If you feel that the correct answer is not accepted for a particular question, do the following:

  • Submit your code to obtain your answer using the “Attach text” button on the right side of the screen. Do not attach a file, use the attach text feature.
  • Submit an issue describing why your answer is correct and reference your attached code.

Questions that are disputed after the exam that do not have the relevant code to support your claim will likely be ignored. (While errors in creating the questions are certainly possible, it is also likely that you simply do not understand the question. It is a better strategy to simply move on to another question than to worry about a single question for a long time.)


The contents below here may no longer be necessary. Because of COVID-19, we have had to significantly alter the process for Exam II. Please only reference the information above.


Dates, Times, and Content

Additional information can be found at the CBTF scheduler:

Exam 00A (Orientation Exam)

  • Start: Tuesday, January 21
  • End: Sunday, February 2

Exam 00A is an optional, but highly recommended 50 minute “exam” that will help you become familiar with the CBTF. It will not count for or against your grade. You will get practice using the scheduler to reserve a time, finding and checking into the CBTF, being guided by the proctors (classroom-style), and taking a PrairieLearn quiz that’s focused on teaching students how to be successful in the CBTF and with PrairieLearn.

Exam 00B (Syllabus Exam)

  • Start: Wednesday, February 5
  • End: Friday, February 7

Exam 00B is a required 50 minute “exam” that will help you become familiar with STAT 432 specifics at the CBTF. In particular, this “exam” will contain a PrairieLearn quiz that includes STAT 432 questions that require R and RStudio to complete. This will be an opportunity to practice using R and RStudio in the CBTF.

While Exam 00B does not appear as part of your grade, failure to complete Exam 00B will have two negative effects on grades:

  • A 10% reduction on Exam 01.
  • Forfeiture of accommodation rights in the event of a technology issue on future exam.

To avoid these deductions, you simply need to obtain a score greater than 0%. (However, you should still try all of the questions! That’s the point!)

If you register for the course on or after the dates for Exam 00B, contact the course instructor immediately and accommodation will be provided if necessary.

Exam 01

  • Start: Wednesday, February 26
  • End: Friday, February 28

Exam 01 will cover the material from Quiz 01 to Quiz 05.


CBTF Policies

Please be aware of the CBTF policies. Failure to read the following links will likely negatively impact your exam experience:

If you encounter a (non-coding) technology issue during the exam, you must file an incident report with a proctor. Without documentation, we cannot (and will not) provide accommodation for technology errors encountered during an exam.

Many technology issues can be solved by simply restarting RStudio.

If you feel that the correct answer is not accepted for a particular question, please submit your code to obtain your answer using the “Attach text” button on the right side of the screen. Questions that are disputed after the exam that do not have the relevant code to support your claim will likely be ignored. (While errors in creating the questions are certainly possible, it is also likely that you simply do not understand the question. It is a better strategy to simply move on to another question than to worry about a single question for a long time.)


Academic Integrity

In short, do not cheat. In addition to the Academic Integrity policies outlined by the CBTF, STAT 432 will hold itself to a higher standard. Any violation will be punished as harshly as possible, which at minimum is a failing grade on the exam.

A non-exhaustive list of academic integrity violations:

You are correct that some of these things may be hard to detect, but in addition to writing the exam in a manner that makes transferring this type of information less useful, we have other methods that you are likely unaware of to detect the above violations.

Be aware that the exam window goes until Friday for Exam 01 and Exam 02. However, this does not mean that you may discuss the exam at that time as there may be makeup exams in rare circumstances. Do not discuss any part of the exams until it is announced that you may do so.


Format

You will be given 1 hour and 50 minutes to complete the exams (Exam 01 and Exam 02) which will be administered through PrairieLearn. You will have access to R and RStudio with a large set of packages installed. (Those seen in the quizzes.) The question format will mirror that seen in the quizzes. Some quiz questions will reappear during the exams.

While quizzes allow for unlimited attempts, on exams you will have multiple, but not unlimited attempts.

If you get a question incorrect, a new version will not be generated, however you will not see the submission hints. Follow-up attempts will have the same version of the problem. (This is different than the quizzes where you always get a new version. If a quiz question is reused for the exam, it may indicate that new variants will use a different seed, but no new variants will be generated on the exam.)

Exam 00B will be extremely useful for understanding the above scoring.

In addition to multiple attempts, there will be more questions available than are needed to obtain 100% on Exam 01 and Exam 02. Point cutoffs to obtain full marks will be indicated in the directions of the exam, which you should read as soon as you begin the exam.


R Packages

The following packages and their dependencies are available for use in the CBTF. (This information is up-to-date as of Exam 01. Additional packages may be added ahead of Exam 02.)


Provided Notes


Tips for Studying